It is now becoming clear that heritable and acquired human disorders arise from a complex interplay between multiple genetic components influenced by environmental factors. Therefore, scanning the genome for genetic associations to a specific disease is hampered by the fact that different individuals with the same disease could have different underlying genetic causes. When associations are found, they are often weak or relate to only one population of individuals sharing a particular genetic history. To alleviate this limitation we need a broader view of the molecular assemblages that specify particular phenotypes and need to develop [unreadable] better systems for the analysis of complex phenotypes. Disease phenotypes, especially for the [unreadable] most significant diseases, like cancer, are often the result of a breakdown in a basic cellular or [unreadable] developmental process. Thus, in the long term, this project aims to build a systems view of the [unreadable] molecular mechanisms driving these basic processes and use this information to inform the [unreadable] analysis of genome wide association studies. To begin this process we will build a systems view [unreadable] of early embryogenesis in mouse and human and connect them to specific phenotypic outcomes [unreadable] underlying basic cell and developmental processes. This approach has been very successful in [unreadable] the model C. elegant, where groups of genes required for specific phenotypes were identified [unreadable] on a genome-wide scale. We will use similar computational and experimental approaches as [unreadable] well as use a compendium of data from several model systems to build a systems view of early [unreadable] embryonic development in mouse and human. Specifically we will work on the following three [unreadable] related aims: 1: Develop a computational framework to predict the global molecular map [unreadable] underlying early embryonic development in mouse and human. 2: Gather high-content [unreadable] phenotypic data during early embryogenesis using RNAi of ~400 mouse genes; digitally encode [unreadable] their complex phenotypic effect and combine with transcriptome data and the computational [unreadable] map built in Aim 1. 3: Develop an open-access Web-environment to navigate and analyze these [unreadable] complex data, including time-lapse recordings of early embryonic phenotypes so that they can [unreadable] be used to analyze genome-wide associations' studies. This work will have direct implications [unreadable] for diagnosis of embryonic viability in a clinical setting and, due to the pleiotropic roles of many [unreadable] genes, will in turn provide insights into complex phenotypes associated with a variety of [unreadable] developmental and metabolic disorders as well as other disease processes like cancer, which [unreadable] involves deregulation of cell proliferation vs. differentiation. [unreadable] [unreadable] [unreadable]